Abstract: Estimation of premorbid intelligence is imperative to detect general cognitive decline and therefore it is important to both research and clinical work. Several estimating approaches were introduced. Qualitative estimation based on intelligence-related background such as educational records and occupational achievements can assess premorbid intelligence comprehensively, however, its objectivity is suspected. On the other hand, objective and quantitative estimation based on regression equation of demographic variables inevitably overestimates or underestimates extreme IQs. Approaches depended on 慼old?test or best performance are more individual-specific while their premise is questioned. Combining different approaches can improve predictive validity.

Key words: premorbid intelligence, validity, demographic variables, 慼old/don抰 hold?